Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A plant vesicle-dendritic cell chimera for enhancing cancer immunotherapy.

Nature communications·2026
Same author

The role of pterostilbene in alleviating inflammatory response and pyroptosis in myocardial ischemia-reperfusion by inhibiting the cGAS-STING pathway.

Cytotechnology·2026
Same author

Prognostic value of the CT-based Suidan scoring system in advanced epithelial ovarian cancer.

Frontiers in oncology·2026
Same author

In Reply: Comment on "Nonsurgical Integrative Treatments for Degenerative Symptomatic Lumbar Spondylolisthesis".

Mayo Clinic proceedings·2026
Same author

The Crude Polysaccharide Derived from <i>Agaricus subrufescens</i> Alleviates Alcoholic Liver Injury.

Foods (Basel, Switzerland)·2026
Same author

Global, regional, and national burden of vision impairment from 1990 to 2021, with projections of prevalence to 2050.

Chinese medical journal·2026

Related Experiment Video

Updated: Sep 3, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K

Application of Convolution Neural Network Algorithm Based on Multicenter ABUS Images in Breast Lesion Detection.

Jianxing Zhang1,2, Xing Tao3, Yanhui Jiang3

  • 1Department of Medical Imaging Center, The First Affiliated Hospital, Jinan University, Guangzhou, China.

Frontiers in Oncology
|July 28, 2022
PubMed
Summary

This study developed a deep learning algorithm for breast lesion detection using automated breast ultrasound (ABUS) images. The AI demonstrated high accuracy, particularly for malignant and BI-RADS 4/5 lesions, though smaller lesions require further improvement.

Keywords:
automatic breast ultrasound (ABUS)breast cancerconvolution neural networkdetectionvalidation data

More Related Videos

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.0K
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

6.9K

Related Experiment Videos

Last Updated: Sep 3, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.0K
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

6.9K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Automated Breast Ultrasound (ABUS) is crucial for breast lesion detection.
  • Deep learning algorithms show promise in improving diagnostic accuracy.
  • Multi-center validation is essential for robust AI performance in clinical settings.

Purpose of the Study:

  • To evaluate a Convolutional Neural Network (CNN) algorithm, based on YOLO v5, for breast lesion detection.
  • To assess the algorithm's performance using multi-center ABUS image data.
  • To analyze detection efficiency across various lesion characteristics and sizes.

Main Methods:

  • Utilized 2,538 ABUS datasets from 741 cases across 7 hospitals (Oct 2016 - Dec 2020).
  • Employed internal (452 volumes) and external (2,086 volumes) validation datasets.
  • Analyzed algorithm performance based on lesion size, BI-RADS category, and malignancy, with varying false positive allowances.

Main Results:

  • Achieved high overall detection rates: 78.1% (internal) and 71.2% (external).
  • Demonstrated superior detection for larger lesions (≥10 mm, 87.4%), malignant nodules (98.1% internal, 98.2% external), and BI-RADS 4/5 lesions (96.5% internal, 95.8% external).
  • Lower detection rates observed for lesions <10 mm and BI-RADS 2/3 categories.

Conclusions:

  • The CNN algorithm exhibits strong detection efficiency for breast lesions in ABUS images, validated across multiple centers.
  • The algorithm is particularly effective for malignant and advanced-stage (BI-RADS 4/5) lesions.
  • Further refinement is needed to improve detection of smaller (<10 mm) and less suspicious (BI-RADS 2/3) breast lesions.